Much of what we know about domestication comes from the archaeological record. Increasing knowledge about both plant domestication and the exploitation of wild species is a result of intensifying awareness among researchers of the need to recover plant remains from excavations through more refined recovery techniques. A great deal of information has been obtained by the use of technique known as flotation. When placed in water, soil from an excavation sinks, whereas organic materials, including plant remains, float to the surface. These can then be skimmed off and examined by scientists for identifiable fragments. Other information may be obtained by studying the stomach contents of well-preserved bodies.

Although archaeologists can easily distinguish some plant species in the wild from those that were domesticated, the domestication of animals is more difficult to discern from archaeological evidence, even though many features distinguish wild from domesticated animals. Unlike their wild counterparts, domesticated cattle and goats produce more milk than their offspring need; this excess is used by humans. Wild sheep do not produce wool, and undomesticated chickens do not lay extra eggs. Unfortunately, however, the animal remains found at archaeological sites often exhibit only subtle differences between wild and domesticated species. Researchers have traditionally considered reduction in jaw or tooth size as an indication of domestication in some species, for example, the pig and dog. Other studies have attempted to identify changes in bone shape and internal structure. Although providing possible insights such approaches are problematic when the diversity within animal species is considered because the particular characteristics used to identify "domesticated"stock may fall within the range found in wild herds.

A different approach to the study of animal domestication is to look for possible human influence on the makeup and distribution of wild animal populations, for example changing ratios in the ages and sexes of the animals killed by humans. Archaeological evidence from Southwest Asia shows that Paleolithic(2.6 million to about 12,000 years ago)hunters, who killed wild goats and sheep as a staple of their lifestyle, initially killed animals of both sexes and of any age. However, as time went on, older males were targeted, whereas females and their young were spared. Some sheep bones dating back 9,000 years have been found in sites in Southwest Asia far away from the animals' habitat, suggesting that animals were captured to be killed when needed.

Observations such as these may suggest human intervention and incipient domestication, but conclusions need to be carefully assessed. Recent research has pointed out that sex ratios and percentages of juvenile individuals vary substantially in wild populations. Moreover, all predators, not just humans, hunt selectively (choose to hunt some animals but leave others alone). Finally, information on the ancient distribution of animal species is unknown.

In the absence of direct evidence from plant and animal remains, archaeologists attempting to examine the origins of food production at times indirectly infer a shift to domestication. For example, because the food-processing requirements associated with food production, as opposed to hunting and gathering, necessitated specific technological innovations, food-processing artifacts such as grinding stones are found more frequently at Neolithic(11,500-5, 500 years ago) than at Paleolithic sites. In addition, Neolithic peoples had to figure out ways to store food crops because agricultural production is seasonal. Thus, during the Neolithic age, structures used as granaries became increasingly common, allowing for the stockpiling of large food supplies against periods of famine. Agricultural peoples constructed large and small granaries or storage bins and pits out of such diverse materials as wood, stone, brick, and clay. Remnants of these storage structures are found archaeologically. Broken pieces of pottery, too, often give clues to Neolithic communities. Whereas nomadic hunter-gatherers could not easily carry heavy clay pots in their search for new herds and food sources, the settled agrarian lifestyle encouraged the development of pottery, which would facilitate the cooking and storing of food.

Generalizations about farming cannot be made solely on the basis of indirect evidence such as pottery, however, as the same artifact inventory is not associated with the transition to domestication in all cultural settings. In many instances, evidence for domestication precedes the use of pottery. For example, in some sites in Southwest Asia, domesticated barley appears before the use of pottery. Conversely, some of the earliest pottery yet discovered, some 10, 500 years old, was produced by the Jomon culture of Japan, sedentary hunting-gat-and- society.